Merge branch 'master' into chore/telemetry-for-deployment-configs

Author: evakravi

.github/workflows/codeql.yml

@@ -0,0 +1,35 @@
+name: "CodeQL"
+on:
+  push:
+    branches: [ "master" ]
+  pull_request:
+    branches: [ "master" ]
+  schedule:
+    - cron: '30 8 * * *'
+jobs:
+  analyze:
+    name: Analyze (${{ matrix.language }})
+    runs-on: ${{ 'ubuntu-latest' }}
+    permissions:
+      security-events: write
+      packages: read
+
+    strategy:
+      matrix:
+        include:
+          - language: python
+            build-mode: none
+          - language: java-kotlin
+            build-mode: none
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@6ccd57f4c5d15bdc2fef309bd9fb6cc9db2ef1c6
+      - name: Initialize CodeQL
+        uses: github/codeql-action/init@4b1d7da102ff94aca014c0245062b1a463356d72
+        with:
+          languages: ${{ matrix.language }}
+          build-mode: ${{ matrix.build-mode }}
+      - name: Perform CodeQL Analysis
+        uses: github/codeql-action/analyze@4b1d7da102ff94aca014c0245062b1a463356d72
+        with:
+          category: "/language:${{matrix.language}}"

src/sagemaker/fw_utils.py

@@ -152,7 +152,6 @@
     "2.1.0",
     "2.1.2",
     "2.2.0",
-    "2.3.0",
     "2.3.1",
 ]
 

src/sagemaker/image_uri_config/huggingface-llm.json

@@ -717,7 +717,7 @@
                 "tag_prefix": "2.3.0-tgi2.2.0",
                 "repository": "huggingface-pytorch-tgi-inference",
                 "container_version": {
-                    "gpu": "cu121-ubuntu22.04"
+                    "gpu": "cu121-ubuntu22.04-v2.0"
                 }
             }
         }

src/sagemaker/image_uri_config/pytorch-smp.json

@@ -8,7 +8,7 @@
             "2.1": "2.1.2",
             "2.2": "2.3.1",
             "2.2.0": "2.3.1",
-            "2.3": "2.4.0"
+            "2.3.1": "2.4.0"
         },
         "versions": {
             "2.0.1": {

tests/integ/test_sklearn.py

@@ -159,8 +159,6 @@ def test_deploy_model(
 def test_deploy_model_with_serverless_inference_config(
     sklearn_training_job,
     sagemaker_session,
-    sklearn_latest_version,
-    sklearn_latest_py_version,
 ):
     endpoint_name = unique_name_from_base("test-sklearn-deploy-model-serverless")
     with timeout_and_delete_endpoint_by_name(endpoint_name, sagemaker_session):
@@ -173,7 +171,7 @@ def test_deploy_model_with_serverless_inference_config(
             model_data,
             ROLE,
             entry_point=script_path,
-            framework_version=sklearn_latest_version,
+            framework_version="1.0-1",
             sagemaker_session=sagemaker_session,
         )
         predictor = model.deploy(

tests/integ/test_xgboost.py

@@ -121,11 +121,9 @@ def test_training_with_network_isolation(
         ]
 
 
-@pytest.mark.skip(reason="re:Invent keynote3 blocker. Revisit after release")
 def test_xgboost_serverless_inference(
     xgboost_training_job,
     sagemaker_session,
-    xgboost_latest_version,
 ):
     endpoint_name = unique_name_from_base("test-xgboost-deploy-model-serverless")
     with timeout_and_delete_endpoint_by_name(endpoint_name, sagemaker_session):
@@ -139,7 +137,7 @@ def test_xgboost_serverless_inference(
             model_data=model_data,
             role=ROLE,
             entry_point=os.path.join(DATA_DIR, "xgboost_abalone", "abalone.py"),
-            framework_version=xgboost_latest_version,
+            framework_version="1.5-1",
         )
 
         xgboost.deploy(

tests/unit/sagemaker/image_uris/test_huggingface_llm.py

@@ -42,7 +42,7 @@
         "2.0.0": "2.1.1-tgi2.0.0-gpu-py310-cu121-ubuntu22.04",
         "2.0.1": "2.1.1-tgi2.0.1-gpu-py310-cu121-ubuntu22.04",
         "2.0.2": "2.3.0-tgi2.0.2-gpu-py310-cu121-ubuntu22.04",
-        "2.2.0": "2.3.0-tgi2.2.0-gpu-py310-cu121-ubuntu22.04",
+        "2.2.0": "2.3.0-tgi2.2.0-gpu-py310-cu121-ubuntu22.04-v2.0",
     },
     "inf2": {
         "0.0.16": "1.13.1-optimum0.0.16-neuronx-py310-ubuntu22.04",

tests/unit/sagemaker/image_uris/test_smp_v2.py

@@ -27,6 +27,7 @@ def test_smp_v2(load_config):
         "torch_distributed": {"enabled": True},
         "smdistributed": {"modelparallel": {"enabled": True}},
     }
+
     for processor in PROCESSORS:
         for version in VERSIONS:
             ACCOUNTS = load_config["training"]["versions"][version]["registries"]
@@ -38,6 +39,11 @@ def test_smp_v2(load_config):
                         if "2.1" in version or "2.2" in version or "2.3" in version:
                             cuda_vers = "cu121"
 
+                        if "2.3.1" == version:
+                            py_version = "py311"
+
+                        print(version, py_version)
+
                         uri = image_uris.get_training_image_uri(
                             region,
                             framework="pytorch",